Feature space reformation-based teacher-student architecture noise-tag-containing image classification method
A technology of feature space and classification method, applied in neural architecture, neural learning method, biological neural network model, etc., can solve the problem of ignoring the benefits of mutual learning, to promote self-learning, improve performance, robustness and effectiveness excellent effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0027] 1, related work
[0028] 1.1 Noise Label
[0029]Deep neural networks can easily adapt to any noise due to its powerful learning ability. Training depth neural networks on noise data sets, making network fitting noise samples, greatly reducing the generalization of models. In order to overcome this problem, some researchers have modified the loss function to make the model more robust to noise. This method has great theoretical support, but with the increase in noise complexity, the effectiveness of the method gradually decreases. At the same time, the modification of the loss function increases the amount of calculation required for training convergence. Since the direct learning noise tag cannot achieve good results, the researchers turn to correct the noise label. Reed et al. [Reed, Scott, et al. * # * Training Deep Neural Networks on noisylabels with bootstrapping. * # * Arxiv Preprint Arxiv: 1412.6596 (2014).] Combined the original label with model prediction to genera...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap